The objective of this project is to investigate the involvement of coherent amplitude modulation (AM) in sound source determination. Due to the nature of many sound sources, the spectral components that originate from a single source often share a common or coherent AM; the amplitudes of the components of s single sound source vary in a similar manner over time. The aim of the project is to investigate the role that this coherent modulation many play in the fusion and segregation of auditory images as a fundamental ingredient in sound source determination. This proposal has three parts with the major effort directed towards human psychophysics. However, the overall project is a multi-disciplinary study of modulation processing. Therefore, the proposal contains work involving animal (chinchilla) psychophysics and auditory physiology at the level of the cochlear nucleus. Three objectives are to be pursued in the human psychophysical experiments: the first objective is concerned with the manner in which modulation information from specific audio-frequency regions is combined, the second objection evaluates selectivity with masking and discrimination procedures that require listeners to process envelope coherence among specific spectral components in a complex sound, and the third objective is to investigate auditory envelope detection and temporal pattern processing to provide a basis for modeling selective cross-spectral processing of AM. The effects measured with human listeners in the past [most notably Modulation Detection Interference (MDI)] and many of those planned during this grant period will be investigated in a mammalian animal model, the chinchilla. The overall aim of the physiology section of the project is to understand AM processing primarily at the level of the cochlear nucleus in the chinchilla, especially for complex, low-rate modulation patterns. As in the other projects, this three-pronged approach is viewed as important for understanding information processing by the auditory system.